Cancer Immunotherapy
Most cancer patients are treated with some combination of surgery, radiation, and chemotherapy.
Radiation and chemotherapy have the disadvantage of destroying healthy as well as malignant
cells and thus can cause severe side-effects.
What is needed are more precisely-targeted therapies.
One long-held dream is that the specificity of immune mechanisms could be harnessed against
tumor cells. This might use


the patient's own immune system (immunostimulants)
the transfer of
o antibodies (i.e. passive immunization) or
o T cells from an outside source.
Ideally, these agents would be targeted to molecules expressed on the cancer cells but not on
healthy cells. However, such tumor-specific antigens have been hard to find, and so many of
the immune agents now in use do target healthy cells as well.
1. Immunostimulants
There is considerable evidence that cancer patients have T cells that are capable of attacking
their tumor cells. In fact, it may be that the appearance of cancer is a failure of immune
surveillance: the ability of one's own immune system to destroy cancer cells as soon as they
appear. [The evidence for immune surveillance is discussed on a separate page — link to it.] But
what to do if they fail?
Immunostimulants are nonspecific agents that tune-up the body's immune defenses. There have
been some successes with




injecting adjuvant-like agents directly into the tumor. The only one that succeeds often
enough to remain in use is the bacterial preparation BCG. Introduced into the bladder, it
can help eradicate early-stage bladder tumors.
Oral therapy with levamisole, a drug widely-used for deworming (people as well as
animals), has been used to treat a variety of cancers but with inconsistent results.
interleukin-2 (IL-2), a potent growth factor for T cells;
alpha-interferon (IFN-α)
2. Cancer Therapy with Monoclonal Antibodies
A number of monoclonal antibodies show promise against cancer, especially cancers of white
blood cells (leukemias, lymphomas, and multiple myeloma).
Link to a discussion of how monoclonal antibodies are made.
Some examples:







Rituximab (trade name = Rituxan®). Used to treat B-cell lymphomas. The CD20
molecule to which it binds is present on most B-cells, healthy as well as malignant, but
over the months following treatment, new healthy B cells are formed from precursors that
do not have CD20 and thus were not destroyed by the treatment.
Herceptin®. Binds HER2, a growth factor receptor found on some tumor cells (some
breast cancers, lymphomas). The only monoclonal so far that seems to be effective
against solid tumors.
Alemtuzumab (MabCampath®). Binds to CD52, a molecule found on white blood cells.
Has produced remission of chronic lymphocytic leukemia.
Lym-1 (Oncolym®). Binds to the HLA-DR-encoded histocompatibility antigen that can
be expressed at high levels on lymphoma cells.
Bevacizumab (Avastin®). Binds to vascular endothelial growth factor (VEGF) thus
blocking its action and depriving the tumor of its blood supply.
Cetuximab (Erbitux®). Used to treat colorectal cancers.
Ipilimumab (Yervoy®). Blocks the immunosuppresive effects of CTLA-4 (thus a
double-negative). So unlike the other monoclonals listed here, it acts as an
immunostimulant. Approved by the U.S. Food and Drug Administration on 25 March
2011 for use against metastatic melanoma. Because it works to enhance the body's
immune response, e.g., by cytotoxic T lymphocytes (CTL), it may well turn out to be
useful against other cancers as well.
Immunotoxins
A major problem with chemotherapy is the damage the drugs cause to all tissues where rapid cell
division is going on. What is needed is a "magic bullet", a method of delivering a cytotoxic drug
directly and specifically to tumor cells, sparing healthy cells. Such a magic bullet would have
two parts:


a monoclonal antibody specific for the cancer cell attached to
a cytotoxic drug or toxin that kills the cell once it gets inside.
Some two dozen immunotoxins are in clinical trials. One that has already received FDA
approval:
Adcetris®. A conjugate of


a monoclonal antibody that binds CD30, a cell-surface molecule expressed by the cells of
some lymphomas but not found on the normal stem cells needed to repopulate the bone
marrow.
vedotin, a drug that blocks mitosis by preventing the polymerization of tubulin (needed
to form the mitotic spindle).
The vedotin is attached to the monoclonal antibody by a bridge that is cleaved once the conjugate
is safely inside the tumor cell releasing the toxin to do its work there.
In one trial, 73% of the patients with Hodgkin's lymphoma went into remission.
Radioimmunotherapy
Monoclonal antibodies against tumor antigens can also be coupled to radioactive atoms.
The goal with these agents is to limit the destructive power of radiation to those cells (cancerous)
that have been "fingered" by the attached monoclonal antibody.
Examples:

Zevalin®. This is a monoclonal antibody against the CD20 molecule on B cells (and
lymphomas) conjugated to either
o the radioactive isotope indium-111 (111In) or
o the radioactive isotope yttrium-90 (90Y)
Both are given to the lymphoma patient, the 111In version first followed by the 90Y
version (in each case supplemented with Rituxan®).

Bexxar® (tositumomab). This is a conjugate of a monoclonal antibody against CD20 and
the radioactive isotope iodine-131 (131I). It, too, is designed as a treatment for lymphoma.
Although both Bexxar® and Zevalin® kill normal B cells, they don't harm the B-cell
precursors because these do not express CD20. So, in time, the precursors can repopulate
the body with healthy B cells.
On 3 February 2005, the New England Journal of Medicine reported that 59% of patients
with a B-cell lymphoma were disease-free 5 years after a single treatment with 131Itositumomab (a treatment that was relatively free of the nasty side-effects, e.g., hair loss,
of conventional chemotherapy).
3. Cancer Therapy with T Cells
Tumor destruction is done by cells. Antibodies may help, but only by identifying the cells to be
destroyed, e.g., by macrophages.
But T cells, e.g., cytotoxic T lymphocytes (CTL), are designed to destroy target cells. What
about enlisting them in the fight?
Using a Bispecific Monoclonal Antibody
Blinatumomab is a synthetic monoclonal antibody each arm of which carries a binding site with
a different specificity:

one arm binds to CD19, an antigen found on the surface of B cells and B-cell
lymphomas;

the other arm binds to CD3, a cell-surface molecule on T cells, including cytotoxic T
lymphocytes (CTLs).
By forming a bridge between CD3 and CD19, blinatumomab is able to attach T cells to B cells
and activate the T cells to kill the B cells.
Early clinical trials with blinatumomab on a small number of patients appear quite promising.
Modest doses of the drug produced partial and, in a few cases, complete regression of their
lymphoma.
Allografts of T cells
After total destruction of the patient's own white blood cells ("myeloablative
conditioning")
One approach to curing leukemia (and some other cancers) is to treat the patient with such high
doses of chemotherapy and radiation that not only are the leukemic cells killed, but the patient's
bone marrow is destroyed. If the patient is to survive the treatment, he or she must be given a
transplant of hematopoietic stem cells — the cells from which all blood cells are formed.
Link to discussion of hematopoietic stem cell transplants.
The stem cells can be


an autograft; that is, from bone marrow harvested from the patient and stored before
treatment begins. In this case, however, the marrow must also be treated to purge it of all
cancer cells it may contain before it is returned to the patient. This sometimes fails.
an allograft; that is, cells harvested from another person, usually a family member
sharing the same major histocompatibility molecules.
Allografted hematopoietic stem cells also sometimes fail to cure, but in that case it is because not
all of the patient's leukemic cells were destroyed. However, an infusion of T lymphocytes from
the blood of the same donor that provided the cells can finish off the job.
This effect is called the graft-versus-leukemia effect.
However, most (if not all) of the donor T cells are probably attacking normal cell surface
molecules, not tumor-specific ones. (Even if the donor and recipient are matched for the major
histocompatibility molecules, there will be minor ones that elicit a rejection response.)
So the patient may also suffer life-threatening graft-versus-host disease (GVHD).
After nonmyeloablative conditioning
The graft-versus-leukemia effect lays the foundation for an approach that has shown
considerable promise against various blood cancers and even some solid (e.g., kidney) tumors.






The patient is treated to kill some — but not all — of the bone marrow cells.
Instead of using high doses of radiation to the entire body and chemotherapy, only the
lymphoid organs (spleen, thymus, lymph nodes) are irradiated (called "total lymphoid
irradiation").
Antithymocyte globulin can also be given.
Even though this leaves some cancer cells, it makes it possible for allogeneic bone
marrow stem cells to take up long-term residence in the recipient (just as
immunosuppression allows kidney transplants, etc. to avoid rejection by the recipient —
Link).
This is followed by an infusion of T cells from the same donor. These can then go to
work against the cancer cells without being threatened with rejection by the host.
Once again, though, they will also attack normal cells of the recipient usually causing
graft-versus-host disease (GVHD). However, this promises to be milder than that
following myeloablative conditioning — perhaps because repeated small doses of
radiation favors the survival of natural killer (NK) cells, and these appear to protect
against GVHD.
In mice, the graft-versus-leukemia effect can be enjoyed without the downside of GVHD by
including extra-large numbers of regulatory T cells (Tr cells) in the bone marrow infusion.
Whether this approach could be helpful for humans remains to be seen.
Autografts of T cells: Tumor-Infiltrating Lymphocytes (TIL)
Solid tumors contain lymphocytes that are specific for tumor antigens. For many years, Steven
A. Rosenberg and his associates at the U. S. National Cancer Institute have tried to enlist these
cells in cancer therapy.
On September 19, 2002, he reported his most promising results at that time. The procedure:





Isolate T cells — both CD4+ T-helper cells and CD8+ cytotoxic T lymphocytes (CTL)
from samples of the tumor (melanoma)
Test them in vitro to find the most efficient killers of the melanoma cells.
Grow large numbers of them in culture (using the powerful T-cell growth factor IL-2).
Treat the patient with modest doses of cytotoxic drugs to reduce — but not destroy — the
bone marrow.
Reintroduce the mix of Th cells (CD4+) and CTL (CD8+) into the patient (along with IL2).
The results:


The infused cells usually took up long-term residence.
In 10 of 13 patients, their melanoma cells — including all metastases — regressed either
partially or completely.
In a few cases, the TIL seemed to be reacting to tumor-specific antigens, but in most the target
seems to have been antigens expressed by all melanin-containing cells. Evidence:


Four patients lost normal melanocytes from their skin leaving white patches.
One patient developed inflammation of the uvea, the coat of melanin-containing cells
within the eye.
Gene Therapy: Autografts of genetically-modified T cells
On April 20, 2006, the Rosenberg group reported some success with melanoma patients using a
modification of the TIL procedure.




The patient's T cells were removed and treated with a retroviral vector containing the αβ
TCR specific for a melanoma antigen.
Large numbers of these were grown in culture.
After partial myeloablative conditioning to "make room" for them, the geneticallymodified lymphocytes were infused into the patient.
This application of gene therapy succeeded in eliminating the metastases and providing a
disease-free period of two years in two patients.
See also, Monoclonal T-cell Receptors.
Another example.
The 10 August 2011 online version of the New England Journal of Medicine carried a report by
Porter, D., et al. on their results with one (of three) patients treated for chronic lymphocytic
leukemia (CLL) with an infusion of his own genetically-modified T cells.
The patient's malignant B cells expressed the surface antigen CD19 just as normal B cells do.
T cells were harvested from his blood and later treated with a vector encoding the antigenbinding site of an anti-CD19 antibody along with two other costimulatory molecules. The result:
some 5% of these T cells expressed this synthetic antibody and were activated when they bound
CD19 with it (rather than with their T cell receptor (TCR) which they would normally use).
Injected back into the patient, they proliferated by some 1000-fold and persisted for months.
During this period, they eliminated all his malignant B cells (as well as his normal B cells).
At the time of the report (10 months after treatment), he continued to be free of his cancer.
Lacking normal B cells as well, he needed periodic infusions of immune globulin to keep
infections at bay.
Autografts of a clone of tumor-antigen-specific T cells
The 19 June 2008 issue of the New England Journal of Medicine (Naomi Hunder et al) carried a
report describing the successful treatment of a man with metastasized melanoma using his own T
cells. The procedure:




His leukocytes were harvested and a mixed culture was prepared containing
o antigen-presenting dendritic cells.
o a peptide from the antigen NY-ESO-1. NY-ESO-1 is a protein that is produced by
several types of tumors (e.g., melanoma, lung and breast cancers) but is not
expressed by normal cells (except those in the testis).
o The patient's own T cells.
After repeated stimulation with the antigen, responding cells were cloned by limiting
dilution.
One (of four) antigen-reactive cells was then expanded in culture until
5 billion (5 x 109) identical anti-NY-ESO-1 CD4+ T cells were available to infuse into the
patient.
The result: complete regression of each metastatic clump of melanoma cells, and the patient has
remained free of this lethal cancer for two years since this treatment.
"One swallow does not make a summer", but this result does give hope that in time
immunotherapy will become an effective weapon against cancer.
4. Cancer Vaccines
Any response of the patient's own immune system – immune surveillance – has clearly failed in
cancer patients. The purpose of cancer vaccines is to elicit a more powerful active immunity in
the patient. Several approaches are being explored.
Patient-Specific Cancer Vaccines
Dendritic-Cell Vaccines
Dendritic cells are the most potent antigen-presenting cells. They engulf antigen, process it into
peptides, and "present" these to T cells.
To make a dendritic-cell vaccine,




Harvest dendritic cells from the patient.
Expose these in vitro to antigens associated with the type of tumor in the patient.
o The antigens are found in normal – as well as cancerous – cells of that tissue (e.g.,
tyrosinase in melanocytes, prostatic acid phosphatase [PAP] in prostate cells).
o They may be fused with a stimulatory molecule such as granulocyte-macrophage
colony-stimulating factor (GM-CSF)
Inject these "pulsed" dendritic cells back into the patient.
Hope that they elicit an strong cell-mediated immune response, e.g. by cytotoxic T
lymphocytes (CTL).
On 29 April 2010 the U.S. Food and Drug Administration approved the first anti-cancer vaccine:
a patient-specific dendritic-cell vaccine for use against advanced prostate cancer. The vaccine,
called Provenge®, is produced by pulsing the patient's dendritic cells with a fusion protein
coupling prostatic acid phosphatase [PAP] with GM-CSF.
Patient-Specific Tumor Antigen Vaccines
The antigens in these vaccines are taken from the patient's own tumor cells.



Harvest some tumor cells from the patient.
Ship them to a company that will use them to make complexes with adjuvant materials.
The complexes are returned to be injected into the patient.
Several of such vaccines are currently in clinical trials. An example.
Tumor-Antigen-Specific Vaccines
These vaccines are used to immunize the patient with an antigen universally expressed by tumors
of that type (but not by normal cells) mixed with some form of adjuvant that will enhance the
response.
Examples:


Many cancer patients mount an immune response — both antibody-mediated and cellmediated — against the tumor (and testis) antigen NY-ESO-1. Deliberate immunization
with this protein (plus an adjuvant) boosts this response and has shown some promise in
early clinical trials. (Cells in the testis do not express HLA antigens, so are not at risk
from attack by NY-ESO-1-specific cytotoxic T lymphocytes). [More]
MAGE-A3 is another protein common on cancer cells. A vaccine using MAGE-A3 —
along with an adjuvant — is in Phase III clinical trials to assess its effectiveness against
melanoma and lung cancer.
Unlike patient-specific vaccines, these vaccines can be mass-produced for use in anyone with the
appropriate tumor.
5. Combining Procedures #3 and #4
While tumors are immunogenic in the patient who carries them, they are only weakly so. In the
hopes of improving cancer immunotherapy, clinical trials are now proceeding to test the efficacy
of combining


potent patient-specific cancer immunization with
treating the patient with large numbers of cultured cancer-antigen-specific T cells that
result.
The procedure:



repeatedly immunize the patient with his or her own cancer cells along with a strong
adjuvant (e.g., GM-CSF) followed by
harvesting the patient's leukocytes and growing large numbers of them in the laboratory
before
infusing them into the patient along with interleukin-2.
This combined approach — which generates large numbers of patient-cancer-specific killer T
cells — has been tested against kidney and one type of brain cancer with promising results.
3 September 2011